JP2016522037A - Automatic quality assessment of physiological signals - Google Patents

Abstract

A method and system for receiving a physiological signal from a sensor device associated with a portable device is provided. A qualitative analysis may be performed on each noise source of the plurality of noise sources in the physiological signal to determine a corresponding plurality of qualitative ratings. Also, at least the plurality of qualitative ratings may be used to determine whether to report the physiological signal to a remote location. In one example, a quantitative analysis is performed on each noise source of the plurality of noise sources to determine an overall quality level and to determine whether to report the physiological signal to the remote location also using the overall quality level.

Description

  Embodiments generally relate to health monitoring. More specifically, embodiments relate to automatic quality assessment of physiological signals in a home health care device.

  Traditionally, health care involves obtaining physiological signals such as electrocardiograph (ECG) readings from individuals in clinical settings such as hospitals, clinics, and other medical centers. There is. In such cases, in order to make health decisions, medical professionals connect various sensors to individuals / patients and interpret readings. If a medical professional determines that the ECG measurements are not reliable or of an appropriate quality, he or she adjusts the detection settings and environment before making health decisions based on these measurements. However, for home use, the patient has the medical and / or technical knowledge needed to identify unreliable or poor quality measurements and make appropriate adjustments to the detection settings / environment. Often does not have. As a result, sub-optimal health care (eg, inappropriate diagnosis, increased costs, and / or increased patient risk) may occur.

Various advantages of the embodiments will become apparent to those skilled in the art after reading the following specification and claims with reference to the accompanying drawings.
It is a figure which shows an example of the signal set by one Embodiment. It is a figure which shows the example of the detection setting by embodiment. It is a figure which shows the example of the detection setting by embodiment. It is a flowchart which shows an example of the evaluation method of the physiological signal by one Embodiment. 6 is a graph illustrating an example of a weighted approach according to one embodiment. FIG. 2 is a block diagram illustrating an example of a logical architecture according to one embodiment. 2 is a block diagram illustrating an example platform according to one embodiment. FIG.

  FIG. 1 is a diagram showing a plurality of signals related to individual / patient management in a home health management apparatus. In the illustrated example, the physiological signal 10 is an electrocardiograph (ECG) signal, for example, and may be considered reliable because it is free of noise. Although the exemplary physiological signal 10 includes ECG information, in other examples, the physiological signal 10 includes blood pressure information, pulse oximeter information, electroencephalograph (EEG) information, photoelectric volume pulse wave recorder (PPG) information, and the like. May be included.

  Depending on the detection settings and / or environment, multiple noise sources 12 (12a-12e) are superimposed on a physiological signal, such as signal 10, to reduce the quality and / or reliability of these signals. For example, the power source (for example, 50/60 Hz) interference source 12a is generated by a nearby low-frequency electric device or a power source line of a building. The muscle noise source 12b is generated by involuntary muscle contraction of the patient due to anxiety or the like, and the motion artifact noise source 12c is generated by patient movement. In addition, the electromagnetic interference (EMI) source 12d may come from nearby high frequency devices such as mobile phones and other electronic devices, and the baseline wander noise source 12e is between the skin and the electrodes. It can also arise from chemical reactions or other contributors that change the impedance. Thus, each noise source 12 can adversely affect the quality of the measured physiological signal as long as there is a respective type of noise in the physiological signal. Indeed, the noise source 12 can be a particular problem in home health settings due to the relative lack of general patient medical and / or technical knowledge.

  As will be described later in detail, both the qualitative analysis and the quantitative analysis may be performed in the home health care apparatus of each noise source 12. These analyzes may be used to determine when to report a physiological signal to a remote location, such as a clinical health care facility (eg, a hospital, clinic, or other medical center). This analysis may also be used to guide the patient when modifying sensor settings and / or environments to increase the reliability of the reported physiological signal.

  Here, referring to FIG. 2A, a home health care environment is shown, and the patient 14 uses the portable device 16, for example, ECG measurement value, blood pressure measurement value, pulse oximeter measurement value, EEG measurement value, PPG measurement. Take measurements such as values. In the illustrated example, the portable device 16 has one or more sensors (eg, electrodes, contacts) that can be crimped to a body part (eg, chest, arm, head, etc.) of the patient 14 to measure the physiological state of the patient 14. Etc.). The portable device 16 generates one or more physiological signals along with the measurements. The physiological signal may be transmitted to the health care network 20. As will be described in more detail later, the portable device 16 automatically performs a quality assessment before sending a physiological signal to the health care network 20, and the assessment indicates that the previous measurement is unreliable. Sometimes, it may be configured to guide the patient 14 when taking additional measurements.

  The health care network 20 may then provide the reported physiological signal to medical personnel such as physicians, nurses, and clinicians. Medical personnel may also send advice to the patient 14 via the health management network 20 and / or the portable device 16. In addition to the built-in sensor 18, the mobile device 16 is a wireless smart phone having other functions such as message transmission / reception (for example, text message transmission / reception, instant message transmission / reception / IM, e-mail, etc.), calculation, media playback, It may be a computing platform such as a smart tablet, personal digital assistant (PDA), mobile internet device (MID), notebook computer, or convertible tablet.

  FIG. 2B shows a home health care environment in which the patient 14 uses the measurement accessory 21 and the portable device 22 to, for example, ECG measurement value, blood pressure measurement value, pulse oximeter measurement value, EEG measurement value, PPG measurement value, etc. Take the measured value. In the illustrated example, the measurement accessory 21 includes one or more sensors (eg, electrodes, contacts, etc.) that can be crimped to a body part of the patient 14 to measure the physiological state of the patient 14. The exemplary measurement accessory 21 generates one or more physiological signals along with the measurements. The physiological signal may be transmitted to the portable device 22. As in the case of the portable device 16 (FIG. 2A), the measurement accessory 21 or the portable device 22 automatically evaluates its quality before sending the physiological signal to the health care network 20, and the previous measured value. May be configured to guide the patient 14 when taking additional measurements.

  As already explained, the health management network 20 provides the reported physiological signals to the medical professional. The medical professional may send advice to the patient 14 via the health management network 20 and / or the portable device 22. The mobile device 22 may be a computing platform such as a wireless smart phone, smart tablet, PDA, MID, notebook computer, convertible tablet, etc. that has message transmission / reception, computation, media playback and / or other functions.

  FIG. 3 illustrates a method 26 for evaluating physiological signals in a home health care device. The method 26 can be performed as a set of logical instructions stored on a machine or computer readable storage medium such as, for example, random access memory (RAM), read only memory (ROM), programmable ROM (PROM), firmware, flash memory, etc. In software, for example, configurable logic such as a programmable logic array (PLA), field programmable gate array (FPGA), complex programmable logic device (CPLD), eg, special purpose integrated circuit (ASIC), complementary metal oxide semiconductor (CMOS) or fixed function logic hardware using circuit technology such as transistor transistor logic (TTL) technology, or any combination thereof. For example, the computer program code for executing the operation shown in the method 26 may be an object-oriented programming language such as Java (registered trademark), Smalltalk, C ++, a conventional procedural programming language such as a C programming language or a similar programming language, Can be written in any combination of one or more programming languages including:

  In exemplary processing block 28, a physiological signal is received from a sensor device associated with the portable device. As already noted, the physiological signal may be related to ECG measurements, blood pressure measurements, pulse oximeter measurements, EEG measurements, and the like. The signal strength of the physiological signal may be estimated at block 30. In one example, the estimation of signal strength includes de-noising of a physiological signal using a multiband filter. The filter used in the denoising procedure may take into account the frequency profile of various types of noise sources (eg, power supply interference, muscle noise, motion artifact noise, EMI, baseline wonder noise, etc.). Block 30 may also include signal processing that identifies one or more reference points in the filtered signal. For example, the reference point in the ECG signal may correspond to the R wave of the ECG signal (that is, upward deviation in the QRS complex). Thus, the reference point is used to calculate the signal strength of the physiological signal.

  In block 32, a first noise source may be extracted from the physiological signal. For example, in the case of the above-described baseline wonder noise 12e (FIG. 1), in block 32, a digital low-pass filter (LPF) having a cutoff frequency of 1 Hz, a cubic spline, or the like may be applied to the physiological signal. Alternatively, the denoised physiological signal obtained at block 30 may be subtracted from the signal at block 32 to separate baseline wonder noise 12e from the physiological signal. In this example, the output of block 32 may be only the baseline wander extracted from the physiological signal. In exemplary block 34, noise estimation of the first noise source is performed. For example, in the case of baseline wander noise 12e (FIG. 1), the noise estimate does not accept outlier data in the isolated baseline wander noise and determines the area under the resulting curve / Calculation may be included. The area under the noise curve is particularly effective for low frequency noise such as baseline wander noise. At block 34, the estimated noise related to the strength of the physiological signal estimated at block 30 may be normalized.

  In block 36, a quantitative analysis may be performed on the first noise source. More specifically, an individual qualitative rating QRi (eg, “Good”, “Fair”, “Poor”) is compared with the first noise by comparing the estimated noise of the first noise source with an appropriate threshold. May be assigned to the source. In this context, medical professionals typically make visual assessments of physiological signals to determine acceptable quality, so a qualitative threshold that matches the visual sharpness / evaluation of a manual is determined. You may choose. For example, the rating criteria may be implemented as shown in Table 1 below.

例示のブロック３８において、第１のノイズ源に対する定量的分析を行う。より具体的には、第１のノイズ源の信号対ノイズ比（ＳＮＲｉ）は、ブロック３０の推定された生理学的信号強度と、ブロック３４の推定及び規格化されたノイズとに基づいて計算してもよい。後でより詳細に説明するように、生理学的信号の全体的品質レベルを求めるため、第１のノイズ源のＳＮＲｉを後で他のノイズ源のＳＮＲと合成してもよい。

In exemplary block 38, a quantitative analysis is performed on the first noise source. More specifically, the signal-to-noise ratio (SNRi) of the first noise source is calculated based on the estimated physiological signal strength of block 30 and the estimated and normalized noise of block 34. Also good. As will be explained in more detail later, the SNRi of the first noise source may later be combined with the SNRs of other noise sources to determine the overall quality level of the physiological signal.

  Similarly, at block 40, a second noise source may be extracted from the physiological signal. For example, in the case of the power main interference source 12a (FIG. 1), in block 40, a digital elliptic bandpass filter (BPF) having a center frequency of 50 Hz or 60 Hz, wavelet transform, or the like is applied to a physiological signal. May be. Alternatively, the denoised physiological signal obtained at block 30 may be subtracted from the filtered signal at block 40 to separate the power source interferer from the physiological signal. In this example, the output of block 40 may be only 50 or 60 Hz power supply noise extracted from the physiological signal. In exemplary block 42, noise estimation of the second noise source is performed. For example, in the case of the power supply interference source 12a (FIG. 1), the noise estimation may include calculating a peak-to-peak average of the noise curve of the separated power main interference. Peak-to-peak averaging may be particularly effective against high frequency noise such as power supply interference. At block 42, the estimated noise related to the strength of the physiological signal estimated at block 30 may be normalized.

  For the first noise source, at block 44 a qualitative analysis is performed on the second noise source. Thus, as described with reference to Table 1, by comparing the estimated noise of the second noise source with an appropriate threshold, an individual qualitative rating QR2 (eg, “Good”, “Fair”, “Poor”) may be assigned to the second noise source. In exemplary block 46, a quantitative analysis is performed on the second noise source. More specifically, the SNR 2 of the second noise source may be calculated based on the estimated physiological signal strength of block 30 and the estimated and normalized noise of block 42.

  An exemplary noise extraction and estimation procedure may be performed for each noise source of the plurality of noise sources of the physiological signal. For example, in the case of muscle noise source 12b (FIG. 1), noise extraction may include applying a digital BPF with a center frequency of 2 Hz or 100 Hz to the physiological signal. In the case of the motion artifact noise source 12c (FIG. 1), for noise extraction, removal of power source interference using a digital notch filter with a center frequency of 50 Hz or 60 Hz, and application of a digital LPF, a cubic spline, etc. with a cutoff frequency of 5 Hz. including. The noise extraction approach can also be tailored to EMI source 12d (FIG. 1) and other types of noise in physiological signals. In each case, the denoised physiological signal obtained at block 30 may be subtracted from the filtered signal to separate a particular type of noise from the physiological signal.

  With respect to noise estimation, relatively high frequency noise, such as muscle noise source 12b and / or EMI source 12d, may be estimated by calculating a peak-to-peak average of the noise curve. On the other hand, noise with relatively low frequency noise, such as the motion artifact noise source 12c, does not accept outliers in the separated noise and may be estimated by determining the area under the resulting curve.

  An exemplary qualitative and quantitative analysis may be performed for each of a plurality of noise sources. In this manner, a plurality of qualitative ratings are obtained, and the plurality of qualitative ratings may correspond to a plurality of noise sources. A plurality of SNRs may be obtained for a plurality of noise sources. Further, after assigning individual qualitative ratings to different noise types, the individual qualitative ratings (QR1, QR2,...) Are synthesized using a scoring function to obtain “Good”, An overall qualitative rating (OQR) for “Fair” or “Poor” may be determined. Thus, the exemplary block 48 determines the OQR of the physiological signal.

  Block 48 may also determine the overall quality level (OQL) of the physiological signal. Here, OQL may be based on both individual qualitative analysis (QR1, QR2,...) And individual quantitative analysis (SNRi, SNR2,...). More specifically, individual SNRs may be combined by a dynamic weighting function to form a single value (eg, in the range of zero to 10). If certain noise types are excessive and the physiological signal quality is at an unacceptable level, the weighting of the weighting function may be changed dynamically. For example, if the ECG signal only has baseline wander noise and the ECG signal is unreliable, the weighting function adjusts the weight of this noise type relative to other noise types and You may carefully consider which of the acceptable mild to moderate noise types is better. Dynamic weight function is OQL

という関数であるように実装してもよい。
ここで、ｎはノイズタイプ数であり、Ｗｋはあるノイズタイプに割り当てられる動的加重であり、ＳＮＲｋはそのノイズタイプの信号対ノイズ比である。このように、上記の式では、ノイズタイプの加重Ｗｋは、対応する個別の定性的レーティング（ＱＲｋ）に基づき動的に変化してもよい。ブロック４８では、ノイズ除去された生理学的信号と、その生理学的信号に関連する定性的及び定量的情報（例えば、個別のＱＲｓ，ＯＱＲ，ＯＱＬなど）とを、後で使うため格納する。

You may implement so that it is a function.
Here, n is the number of noise types, Wk is a dynamic weight assigned to a certain noise type, and SNRk is the signal-to-noise ratio of that noise type. Thus, in the above equation, the noise type weight Wk may dynamically change based on the corresponding individual qualitative rating (QRk). At block 48, the denoised physiological signal and qualitative and quantitative information associated with the physiological signal (eg, individual QRs, OQR, OQL, etc.) are stored for later use.

  FIG. 4 shows a weighting curve 50 that can be used to assign weights to individual noise types. In general, if the qualitative rating of a noise type is Poor, its associated weight may suddenly go to zero so as to significantly reduce the SNR contribution of that noise type to OQL. With such an approach, the presence of excessive noise types can be effectively amplified by greatly reducing OQL.

  Returning to FIG. 3, in block 52, it is determined whether the quality condition is satisfied. The quality condition is, for example, for any noise type, the individual QR is not “Poor”, the OQR is either “Good” or “Fair”, and the OQL is higher than a certain threshold (eg, 5 out of 10), Or any combination thereof. When the quality condition is not met, in exemplary block 54, it is determined whether the maximum number of measurements (eg, 3) has been reached. If not, a user prompt is generated at block 56. One or more additional measurements may be requested at a user prompt (eg, “Please make another ECG measurement”).

  The user prompt may include a recommendation tailored to one or more noise sources among the plurality of noise sources. For example, in the case of baseline wonder noise, the patient is required to hold the device lightly and with uniform pressure. If muscle tremor is excessive, the patient may be asked to relax and support his hands. In the case of motion artifact noise, the patient may still be asked not to move the chest too much when breathing. If power interference or EMI is excessive, the patient may be asked to move or turn off nearby devices. You may make recommendations specific to other noise types. When the patient is prompted, the exemplary method 26 is repeated to determine a plurality of physiological signals associated with the corresponding plurality of measurements, and qualitative and quantitative for each signal of the plurality of physiological signals. Conduct a statistical analysis. The desired physiological signal and associated qualitative and quantitative data may be stored for later use, as described above.

  When the quality condition is met or the maximum number of measurements is reached, the best physiological signal is selected based on the qualitative and quantitative analysis results at block 58 and the best selected at block 60 is illustrated. Physiological signals (and associated qualitative and quantitative data) may be reported to the remote location.

  FIG. 5 shows a logical architecture 51 (51a-51g) for evaluating physiological signals in a home health care device. Illustratively, the sensor interface 51a receives a physiological signal from a sensor device associated with the portable device, and the qualitative module 51b performs a qualitative analysis of each noise source of the plurality of noise sources in the physiological signal and responds accordingly. Find multiple qualitative ratings. The selection module 51c determines whether to report the physiological signal to the remote location using at least a plurality of qualitative ratings.

  In one example, architecture 51 also includes a quantification module 51d that performs a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level. The overall quality level may also be used to determine whether to report the physiological signal to a remote location. More specifically, quantification module 51d assigns weights to signal-to-noise ratios associated with multiple noise sources based on multiple qualitative ratings.

  The qualitative module 51b may also combine a plurality of qualitative ratings to determine an overall qualitative rating, which is used to determine whether to report a physiological signal to a remote location. The example architecture 51 includes a noise extraction module 51e that filters a physiological signal and a noise estimation module 51f that performs noise estimation of the filtered physiological signal. Architecture 51 may also include a user interface (UI) that generates user prompts if quality conditions are met. As described above, the user prompt may require one or more additional measurements and / or may include recommendations tailored to one or more of the plurality of noise sources.

  Referring now to FIG. 6, a computing platform 62 is shown. Platform 62 includes computing functions (eg, PDAs, laptops, smart tablets, etc.), communication functions (eg, wireless smartphones, etc.), imaging functions, media playback functions (eg, smart television / TV), Or it may be part of a device having any combination of the above functions (eg, portable internet device / MID). Illustratively, the platform 62 includes a processor 64, an integrated memory controller (IMC) 66, an input / output (IO) module 68, a system memory 70, a network controller 72, a sensor device 74, a mass storage device 76 (eg, optical disc, hard disk drive). / HDD, flash memory), one or more user interface (UI) devices 82, and a battery 80 that powers the platform 62. The processor 64 may include a core region having one or more processor cores 78.

  The illustrated IO module 68 may be called a south bridge or a south complex in the chip set, but functions as a host controller and communicates with the network controller 72. The network controller 72 provides off-platform communication functions for a wide range of purposes. For this purpose, mobile phones (eg wideband code division multiple access W-CDMA (Universal Mobile Telecommunications System UMTS), CDMA-2000 (IS-856 / IS-2000) etc.), WiFi (wireless fidelity, eg IEEE 802.11-2007, wireless local area network LAN media access control (MAC) and physical layer (PHY) specifications) 4G LTE (4th generation long term evolution), Bluetooth (eg, IEEE 802.15.1-2005, wireless personal Area network), WiMax (eg, IEEE 802.16-2004, LAN / MAN broadband wireless LAN), global positioning system Temu (GPS), spread spectrum (e.g., 900 MHz), and other radio frequency (RF) telephony purposes. The IO module 68 also includes one or more wireless hardware circuit blocks that support such functions. Although the processor 64 and the IO module 68 are illustrated as separate blocks, they may be implemented as a system-on-chip (SoC) on the same semiconductor die.

  The system memory 70 includes, for example, a double data rate (DDR) synchronous dynamic random access memory (SDRAM, eg, DDR3 SDRAM JEDEC standard JESD79-3C, April 2008) module. The module of the system memory 70 is incorporated in, for example, a single inline memory module (SIMM), a dual inline memory module (DIMM), a small outline DIMM (SODIMM), or the like.

  The illustrated core 78 executes logic 84 that evaluates physiological signals in the home health care device described with reference to FIGS. Thus, logic 84 receives a physiological signal from sensor device 74, performs a qualitative analysis on each noise source of the plurality of noise sources in the physiological signal, and determines a corresponding plurality of qualitative ratings, At least a plurality of qualitative ratings are used to determine whether to report the physiological signal to the remote location. Logic 84 performs a quantitative analysis for each noise source of the plurality of noise sources to determine an overall quality level and to determine whether to report the physiological signal to the remote location also using the overall quality level. The patient may be presented with a user prompt for additional measurements via the UI device 82. The UI device 82 may include a display, a speaker, and the like.

<< Additional Notes and Examples >>
Example 1 includes: a portable device for evaluating physiological signals, the portable device having a battery for supplying power to the portable device, a sensor device, and a sensor interface for receiving a physiological signal from the sensor device. apparatus. The portable device performs a qualitative analysis on each noise source of the plurality of noise sources in the physiological signal and obtains a corresponding plurality of qualitative ratings, and at least the plurality of qualitative ratings. And a selection module for determining whether to report the physiological signal to a remote location.

  Example 2 may include the portable device of Example 1, and further includes a quantification module that performs a quantitative analysis on each noise source of the plurality of noise sources to obtain an overall quality level, and the overall quality The level is used to determine whether to report the physiological signal to the remote location.

  Example 3 may include the portable device of Example 2, wherein the quantification module assigns weights to signal-to-noise ratios associated with the plurality of noise sources based on the plurality of qualitative ratings.

  Example 4 may include the portable device of Example 1, wherein the qualitative module combines the plurality of qualitative ratings to determine an overall qualitative rating, and the overall qualitative rating is the physiological signal. Is used to determine whether to report to the remote location.

  Example 5 may include the portable device of Example 1, and further includes a user interface that generates a user prompt when a quality condition is not met, the user prompt requesting one or more additional measurements, Includes recommendations tailored to one or more of the multiple noise sources.

  A sixth embodiment may include the portable device of any of the first to fifth embodiments, and for each noise source of the plurality of noise sources, a noise extraction module that filters the physiological signal and the filtered And a noise estimation module for performing noise estimation of the physiological signal.

  Example 7 may include a device for evaluating a physiological signal, for a sensor interface that receives a physiological signal from a sensor device associated with a portable device, and for each noise source of the plurality of noise sources in the physiological signal. A qualitative module that performs a qualitative analysis to determine a corresponding plurality of qualitative ratings, and a selection module that determines whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings.

  Example 8 may include the apparatus of Example 7, and further includes a quantification module that performs a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, and the overall quality level. Is used to determine whether to report the physiological signal to the remote location.

  Example 9 may include the apparatus of Example 8, wherein the quantification module assigns weights to signal-to-noise ratios associated with the plurality of noise sources based on the plurality of qualitative ratings.

  Example 10 may include the apparatus of Example 7, wherein the qualitative module combines the plurality of qualitative ratings to determine an overall qualitative rating, the overall qualitative rating representing the physiological signal. Used to determine whether to report to the remote location.

  Example 11 may include the apparatus of Example 7, and further includes a user interface that generates a user prompt when a quality condition is not met, the user prompt requesting one or more additional measurements, Recommendations tailored to one or more noise sources.

  Example 12 may include the apparatus of any of Examples 7-11, wherein for each noise source of the plurality of noise sources, a noise extraction module that filters the physiological signal and the filtered And a noise estimation module for performing noise estimation of the physiological signal.

  Example 13 may include the following, a method for evaluating a physiological signal, the step of receiving a physiological signal from a sensor device associated with a portable device, and each noise source of a plurality of noise sources in the physiological signal Performing a qualitative analysis on to determine a corresponding plurality of qualitative ratings and determining whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings.

  Example 14 may include the method of Example 13, and further includes performing a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, the overall quality level Is used to determine whether to report the physiological signal to the remote location.

  Example 15 may include the method of Example 14, and further includes assigning weights to signal to noise ratios associated with the plurality of noise sources based on a plurality of qualitative ratings.

  Example 16 may include the method of Example 13, and further includes combining the plurality of qualitative ratings to determine an overall qualitative rating, the global qualitative rating comprising the physiological signal. Used to determine whether to report to the remote location.

  Example 17 may include the method of Example 13, further comprising generating a user prompt when a quality condition is not met, the user prompt requesting one or more additional measurements, Includes recommendations tailored to one or more of the noise sources.

  Example 18 may include the method of any one of Examples 13 to 17, further comprising: filtering the physiological signal for each noise source of the plurality of noise sources; and Performing noise estimation of the generated physiological signal.

  Example 19 includes the following: at least one computer-readable storage medium having a set of instructions that, when executed by a portable device, causes the portable device to perform any one of the methods of Examples 13-18.

  Example 20 includes the following: an apparatus for evaluating physiological signals, comprising means for performing the method of any one of Examples 13-18.

  Thus, the techniques described herein can automatically assess physiological signal quality by measuring the effects of multiple types of noise. Further, the overall signal quality evaluation may be synthesized regardless of the signal noise type. Further, since the noise type may be extracted and quantified separately, the true reason for the noise can be pointed out to the end user / patient. Such an approach allows the patient to correctly correct the cause of noise in successive measurements. In addition, the dynamic weighting approach biases the analysis results to an excessive effect of one noise source (which is generally unacceptable) rather than a mild / moderate effect of multiple noise sources. Call. This approach can also improve performance by selecting the best quality measure from the remeasurement result set.

  Thus, the likelihood of generating a clinically acceptable physiological signal can be improved. This is because only clinically acceptable quality physiological signals can be sent to the health care network for interpretation by medical professionals. Under the approach described here, the time required for the desired medical advice can be greatly reduced. With this approach, an individual can measure himself / herself at a remote location / home with little or no medical or technical training. In fact, the techniques described herein can be used to meet various risk mitigation requirements associated with medical standards.

  Embodiments of the present invention can be utilized with all types of semiconductor integrated circuit (IC) chips. Examples of these IC chips include, but are not limited to, processors, controllers, chipset components, programmable logic arrays (PLA), memory chips, network chips, system-on-chip (SoC), SSD / NAND controller ASICs, etc. Not. In some drawings, signal conductors are represented by lines. Some lines are different, indicate more configuration signal paths, have a number label indicating the number of configuration signal paths, and / or have arrows at one or more ends to indicate the main information flow direction . However, this should not be interpreted as a limitation. Rather, such additional details are used in conjunction with one or more embodiments to facilitate understanding of the circuit. Any displayed signal line may have additional information, but may have one or more signals traveling in multiple directions, for example, a digital or analog line implemented with a differential pair, It may be implemented with any desired type of signaling, such as fiber optic lines and / or single termination lines.

  The illustrated size / model / value / range may be given, but embodiments of the invention are not so limited. As production technology (such as photolithography) matures over time, it is expected to be able to produce smaller devices. In addition, well-known power / ground connections to IC chips and other components are illustrated, if any, for ease of illustration and description and to avoid obscuring aspects of embodiments of the present invention. Some are not. Further, although the configuration is shown in block diagram form, it does not obscure the embodiments of the present invention, and the specific matters relating to the implementation of such block diagram configurations are highly dependent on the platform on which the embodiments are implemented. That is, it is considered that such specific matters are the jurisdiction of those skilled in the art. Where specific details (e.g., circuitry, etc.) are described to describe embodiments of the invention, it will be understood by those skilled in the art that these specific details are absent or modified. Even so, embodiments of the present invention can be implemented. Thus, the above description should be considered exemplary rather than limiting.

  The term “coupled” here refers to any type of direct or indirect relationship between the components in question, electrical coupling, mechanical coupling, fluid coupling, optical coupling, magneto-optical Applicable to mechanical, electromechanical and other couplings. In addition, terms such as “first” and “second” are used here only for ease of explanation, and have no time meaning unless otherwise specified.

  It will be appreciated by those skilled in the art that the wide variety of embodiments of the present invention can be implemented in a variety of forms. Thus, while embodiments of the invention have been described with reference to specific examples thereof, other modifications will become apparent to those skilled in the art upon study of the drawings, specification, and claims. The scope of the embodiments of the invention is not limited to specific examples.

Claims (26)

A portable device for evaluating physiological signals,
A battery for supplying power to the portable device;
A sensor device;
A sensor interface for receiving a physiological signal from the sensor device;
A qualitative module that performs a qualitative analysis on each noise source of the plurality of noise sources in the physiological signal to determine a corresponding plurality of qualitative ratings;
A selection module that determines whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings.   A quantification module that performs a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, the overall quality level determining whether to report the physiological signal to the remote location; The portable device according to claim 1, wherein the portable device is used.   The portable device of claim 2, wherein the quantification module assigns a weight to a signal-to-noise ratio associated with the plurality of noise sources based on the plurality of qualitative ratings.   The qualitative module combines the plurality of qualitative ratings to determine an overall qualitative rating, the global qualitative rating used to determine whether to report the physiological signal to the remote location. Item 2. The portable device according to Item 1. And a user interface that generates a user prompt when the quality condition is not met, the user prompt requiring one or more additional measurements, and a recommendation tailored to one or more of the plurality of noise sources. including,
The portable device according to claim 1. A noise extraction module that filters the physiological signal for each noise source of the plurality of noise sources;
The portable device according to claim 1, further comprising a noise estimation module that performs noise estimation of the filtered physiological signal. A device for evaluating physiological signals,
A sensor interface for receiving a physiological signal from a sensor device associated with the portable device;
A qualitative module that performs a qualitative analysis on each noise source of the plurality of noise sources in the physiological signal to determine a corresponding plurality of qualitative ratings;
A selection module that determines whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings.   A quantification module that performs a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, the overall quality level determining whether to report the physiological signal to the remote location; The device according to claim 7, wherein   The apparatus of claim 8, wherein the quantification module assigns a weight to a signal to noise ratio associated with the plurality of noise sources based on the plurality of qualitative ratings.   The qualitative module combines the plurality of qualitative ratings to determine an overall qualitative rating, the global qualitative rating used to determine whether to report the physiological signal to the remote location. Item 8. The device according to Item 7.   And a user interface that generates a user prompt when the quality condition is not met, the user prompt requiring one or more additional measurements, and a recommendation tailored to one or more of the plurality of noise sources. The apparatus of claim 7, comprising: A noise extraction module that filters the physiological signal for each noise source of the plurality of noise sources;
12. An apparatus according to any one of claims 7 to 11, further comprising a noise estimation module for performing noise estimation of the filtered physiological signal. A method for evaluating a physiological signal comprising:
Receiving a physiological signal from a sensor device associated with the portable device;
Performing a qualitative analysis on each of the plurality of noise sources in the physiological signal to determine a corresponding plurality of qualitative ratings;
Determining whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings.   Further, the method includes performing a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, the overall quality level determining whether to report the physiological signal to the remote location. The method according to claim 13, wherein the method is used.   The method of claim 14, further comprising assigning a weight to a signal to noise ratio associated with the plurality of noise sources based on a plurality of qualitative ratings.   Further comprising combining the plurality of qualitative ratings to determine an overall qualitative rating, the global qualitative rating being used to determine whether to report the physiological signal to the remote location. Item 14. The method according to Item 13.   And generating a user prompt when the quality condition is not met, the user prompt requesting one or more additional measurements and recommending recommendations to one or more of the plurality of noise sources. 14. The method of claim 13, comprising. Filtering the physiological signal for each noise source of the plurality of noise sources;
Further comprising performing noise estimation of the filtered physiological signal.
The method according to any one of claims 13 to 17. On mobile devices,
Receiving a physiological signal from a sensor device associated with the portable device;
Performing a qualitative analysis for each noise source of the plurality of noise sources in the physiological signal to determine a corresponding plurality of qualitative ratings;
A computer program for determining whether to report the physiological signal to a remote location using at least the plurality of qualitative ratings. Causing the portable device to perform a quantitative analysis on each noise source of the plurality of noise sources to determine an overall quality level, wherein the overall quality level determines whether to report the physiological signal to the remote location; Used to
The computer program according to claim 19. Causing the portable device to assign a weight to a signal-to-noise ratio associated with the plurality of noise sources based on the plurality of qualitative ratings;
The computer program according to claim 20. Causing the portable device to combine the plurality of qualitative ratings to determine an overall qualitative rating, the global qualitative rating being used to determine whether to report the physiological signal to the remote location;
The computer program according to claim 19. Causing the portable device to generate a user prompt when a quality condition is not met, the user prompt requesting one or more additional measurements, and recommending recommendations tailored to one or more of the plurality of noise sources; Including,
The computer program according to claim 19. In the portable device,
Filtering the physiological signal for each noise source of the plurality of noise sources;
Causing noise estimation of the filtered physiological signal;
The computer program according to any one of claims 19 to 23. In the portable device,
Receiving a plurality of physiological signals associated with a corresponding plurality of measurements,
Qualitative analysis is performed on each physiological signal of the plurality of physiological signals;
Selecting a best physiological signal from the plurality of physiological signals based at least in part on the qualitative analysis;
Report the best physiological signal to a remote location;
The computer program according to any one of claims 19 to 23.   A computer-readable storage medium storing the computer program according to any one of claims 19 to 25.

Images